[Solved] High availabilty DHCP problem

[ednt]

Hi,

we are running a master slave configuration with CARP on several interfaces.
According to the DHCPv4 hint I set on the master the split value to 256, because if the master is available,
I only want that the master serves dhcp reuests.
This entry should be only done on the master. (according to the hint)
But if I synchronise the slave, this entries are also set at the slave.

Is this correct???

Btw. I still have the problem of "My state: recover" "Peer state: unknown" on the master for some interfaces.
and "My state: communications-interrupted" "Peer state: normal" on the slave.

At the moment both master and slave serves dhcp requests on different interfaces and during the upgrade from
20.7 to 21.1 the slave dhcp clients lost there network settings.

[ednt]

I looked in the /var/dhcp/etc/dhcpd.conf

And read at ISC:

Legal values are between 0 and 256 inclusive, of which the most reasonable is 128. Note that a value of 0 makes the secondary responsible for all clients and a value of 256 makes the primary responsible for all clients.

So the value 256 on both firewalls is correct.

But the help text is missleading and should be changed:

This value is only used on the primary peer, leave blank on the secondary.

According to ISC
https://kb.isc.org/docs/isc-dhcp-41-manual-pages-dhcpdconf
paragraph 'The split statement'
This is simply wrong.